mlrintermbo: Model-Based Optimization for 'mlr3' Through 'mlrMBO'

The 'mlrMBO' package can ordinarily not be used for optimization within 'mlr3', because of incompatibilities of their respective class systems. 'mlrintermbo' offers a compatibility interface that provides 'mlrMBO' as an 'mlr3tuning' 'Tuner' object, for tuning of machine learning algorithms within 'mlr3', as well as a 'bbotk' 'Optimizer' object for optimization of general objective functions using the 'bbotk' black box optimization framework. The control parameters of 'mlrMBO' are faithfully reproduced as a 'paradox' 'ParamSet'.

Version: 0.5.1-1
Imports: backports, checkmate, data.table, mlr3misc (≥ 0.1.4), paradox, R6, lhs, callr, bbotk, mlr3tuning
Suggests: mlr, ParamHelpers, testthat, rgenoud, DiceKriging, emoa, cmaesr, randomForest, smoof, lgr, mlr3, mlr3learners, mlr3pipelines, mlrMBO, ranger, rpart, mco
Published: 2024-06-07
DOI: 10.32614/CRAN.package.mlrintermbo
Author: Martin Binder [aut, cre]
Maintainer: Martin Binder <developer.mb706 at>
License: LGPL-3
NeedsCompilation: no
Materials: README NEWS
CRAN checks: mlrintermbo results


Reference manual: mlrintermbo.pdf


Package source: mlrintermbo_0.5.1-1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): mlrintermbo_0.5.1-1.tgz, r-oldrel (arm64): mlrintermbo_0.5.1-1.tgz, r-release (x86_64): mlrintermbo_0.5.1-1.tgz, r-oldrel (x86_64): mlrintermbo_0.5.1-1.tgz
Old sources: mlrintermbo archive


Please use the canonical form to link to this page.